Differentially Expressed Genes Network Analysis

Kozo Nishida, Kristina Hanspers and Alex Pico

Contents

The R markdown is available from the pulldown menu for Code at the upper-right, choose “Download Rmd”, or download the Rmd from GitHub.

---

This protocol describes a network analysis workflow in Cytoscape for a set of differentially expressed genes. Points covered:

• Retrieving relevant networks from public databases
• Network functional enrichment analysis
• Integration and visualization of experimental data
• Exporting network visualizations

---

1 Installation

if (!requireNamespace("BiocManager", quietly = TRUE))
  install.packages("BiocManager")

if(!"RCy3" %in% installed.packages())
  BiocManager::install("RCy3")

library(RCy3)

2 Getting started

First, launch Cytoscape and keep it running whenever using RCy3. Confirm that you have everything installed and running:

    cytoscapePing()
    cytoscapeVersionInfo()

3 Setup

If you haven’t already, install the STRINGapp

installApp('stringApp')
installApp('yfileslayoutalgorithms')

4 Background

Ovarian serous cystadenocarcinoma is a type of epithelial ovarian cancer which accounts for ~90% of all ovarian cancers. The data used in this protocol are from The Cancer Genome Atlas, in which multiple subtypes of serous cystadenocarcinoma were identified and characterized by mRNA expression.

We will focus on the differential gene expression between two subtypes, Mesenchymal and Immunoreactive.

For convenience, the data have already been analyzed and pre-filtered, using log fold change value and adjusted p-value.

5 Network Retrieval

Many public databases and multiple Cytoscape apps allow you to retrieve a network or pathway relevant to your data. For this workflow, we will use the STRING app. Some other options include:

• WikiPathways
• NDEx
• GeneMANIA

6 Retrieve Networks from STRING

To identify a relevant network, we will query the STRING database in two different ways:

• Query STRING protein with the list of differentially expressed genes.
• Query STRING disease for ovarian cancer.

6.1 STRING Protein Query: Up-regulated genes

• Read the file containing the list of up-regulated genes, TCGA-Ovarian-MesenvsImmuno_UP.txt.

df <- read.table("https://cytoscape.github.io/cytoscape-tutorials/protocols/data/TCGA-Ovarian-MesenvsImmuno_UP.txt")

• Run this code chunk to run STRING protein query with confidence (score) cutoff of 0.4.

string.cmd = paste('string protein query query="', paste(df$V1, collapse = '\n'), '" cutoff=0.4  species="Homo sapiens"', sep = "")
commandsRun(string.cmd)

• The resulting network will load automatically.

The resulting network contains up-regulated genes recognized by STRING, and interactions between them with an evidence score of 0.4 or greater.

getTableColumnNames('edge')

evidence.score <- getTableColumns('edge', "stringdb::score")
min(evidence.score)

7 Enrichment Analysis Options

Next, we are going to perform enrichment anlaysis uing the STRING app. Note that there are several other options, including:

• clusterProfiler: An R package for ORA and GSEA
• g-Profiler: An enrichment analysis website
• EnrichR: A website that performs enrichment against dozens of ontologies and pathway resources
• ClueGO: Creates and visualizes a functionally grouped network of terms/pathways
• BiNGO: GO overrepresentation analysis on networks
• EnrichmentMap: Visualize the results of gene-set enrichment as a network

8 STRING Enrichment

The STRING app has built-in enrichment analysis functionality, which includes enrichment for GO Process, GO Component, GO Function, InterPro, KEGG Pathways, and PFAM.

• With the STRING results network selected, go to Apps → STRING Enrichment → Retrieve functional enrichment.
• When the enrichment analysis is complete, a new tab titled STRING Enrichment will open in the Table Panel.

The STRING app includes several options for filtering and displaying the enrichment results. The features are all available at the top of the STRING Enrichment tab.

• We are going to filter the table to only show GO Process. At the top left of the STRING enrichment tab, click the filter icon . Select GO Process and check the Remove redundant terms check-box. Then click OK.
• Next, we will add a split donut chart to the nodes representing the top terms by clicking on .
• Explore custom setting via in the top right of the STRING enrichment tab.

9 STRING Protein Query: Down-regulated genes

Repeat the network search, enrichment analysis and visualization for the set of down-regulated genes:

• Query STRING protein with the list of symbols in TCGA-Ovarian-MesenvsImmuno_DOWN.txt.

df <- read.table("https://cytoscape.github.io/cytoscape-tutorials/protocols/data/TCGA-Ovarian-MesenvsImmuno_DOWN.txt")
string.cmd = paste('string protein query query="', paste(df$V1, collapse = '\n'), '" cutoff=0.4  species="Homo sapiens"', sep = "")
commandsRun(string.cmd)

• Perform STRING Enrichment analysis on the resulting network.
• Filter the analysis results for non-redundant GO Process terms only, add split donut charts. To distinguish between the visualizations of up- and down-regulated results, pick a different color palette under Change Color Palette under Settings.

Pro-tip: If you remove the Fill Color mapping from the Style Panel (right-click > Edit > Remove…), set the default to light gray, change the split donut to a Pie Chart in Settings and then try out some of the layouts (see Layout menu), you can end up with network views like this…

10 STRING Disease Query

Now, we will query the STRING disease database to retrieve a network of ovarian cancer associated genes, completely independent of our dataset.

string.cmd = 'string disease query disease="ovarian cancer"'
commandsRun(string.cmd)

This will bring in the top 100 ovarian cancer associated genes connected with a confidence score greater than 0.4. (We did not give them as parameters[100 and 0.4]. This is because the default values are those.)

11 Data integration

Next we will import log fold changes and p-values from our TCGA dataset and use them to create a visualization. Since the network and data use different identifiers, we first have to do some quick identifier mapping. In this case, we will use the gene symbol in the display name column to retrieve Entrez Gene identifiers.

getTableColumnNames('node')

mapped.cols <- mapTableColumn("display name",'Human','HGNC','Entrez Gene')

Here we set Human as species, HGNC as Map from, and Entrez Gene as To.

head(mapped.cols)

• mapped.cols displays a report of how many identifiers were mapped. Make note of this information as it impacts all down-stream analysis; If the mapping was unsuccessful, downstream analysis will be as well. You will notice the new column Entrez Gene in the Node Table.

tail(getTableColumnNames('node'))

• Next, we can load the data (a local copy of TCGA-Ovarian-MesenvsImmuno_data.csv).

df <- read.csv("https://cytoscape.org/cytoscape-tutorials/protocols/data/TCGA-Ovarian-MesenvsImmuno_data.csv")

head(df)

And integrate the data with the network (node) table in Cytoscape. - Key Column for Network should be Entrez Gene. - Gene should be the key of the data(df).

loadTableData(df, data.key.column = "Gene", table = "node", table.key.column = "Entrez Gene")

You will notice two new columns (logFC and FDR.adjusted.Pvalue) in the Node Table.

tail(getTableColumnNames('node'))

12 Visualization

We can now use the integrated data to create a network visualization.

• For node Fill Color, create a continuous mapping for logFC, with the following blue-red gradient.

logFC.table <- getTableColumns('node', "logFC")

logFC.min <- min(logFC.table, na.rm = TRUE)
logFC.max <- max(logFC.table, na.rm = TRUE)
print(logFC.min)
print(logFC.max)

getVisualStyleNames()

setVisualStyle("BioPAX")
setNodeFontSizeDefault(4, style.name = "BioPAX")
setNodeColorMapping("logFC", c(-logratio.max, 0, logratio.max), c('#0000FF', '#FFFFFF', '#FF0000'), style.name = "BioPAX")

• Change the default node Fill Color to light grey.

setNodeColorDefault('#D3D3D3', style.name = "BioPAX")

Pro-tip: If you apply the yFiles Organic layout, you can end up with network views like this… (yFiles Layout Algorithms App does not support any automation. Please select it in Cytoscape Desktop menubar.)

The TCGA found several genes that were commonly mutated in ovarian cancer, so called “cancer drivers”. We can add information about these genes to the network visualization, by changing the visual style of these nodes. Three of the most important drivers are TP53, BRCA1 and BRCA2. We will add a thicker, clored border for these genes in the network.

• Select all three driver genes by

selectNodes(c("TP53", "BRCA1", "BRCA2"), by.col = "display name")

• Add a style bypass for node Border Width (5) and node Border Paint (bright pink). You can create a style bypass by

setNodeBorderWidthBypass(getSelectedNodes(), 5)
setNodeBorderColorBypass(getSelectedNodes(), '#FF007F')

The network will now look like this:

13 Other Analysis Options

• Exploring networks: finding paths, hubs and modules (clusterMaker, MCODE, jActiveModules, NetworkAnalyzer, PathFinder)
• Extending networks with Transcription Factors, miRNAs, etc using CyTargetLinker

14 Exporting Networks

Cytoscape provides a number of ways to export results and visualizations:

• As an image:

exportImage('./differentially-expressed-genes', 'PDF')
exportImage('./differentially-expressed-genes', 'PNG')
exportImage('./differentially-expressed-genes', 'JPEG')
exportImage('./differentially-expressed-genes', 'SVG')
exportImage('./differentially-expressed-genes', 'PS')

• To a public repository:

exportNetworkToNDEx("user", "password", TRUE)

• As a Cytoscape JSON file:

exportNetwork('./differentially-expressed-genes', 'cyjs')